WESTERN AUSTRALIA

DATA QUALITY STATEMENT

In 2012, Western Australia (WA) was responsible for one data collection in relation to preschool programs, the WA August Student Census, which collected data using a Unit Record Level (URL) data collection methodology. The WA August Student Census included an additional component, the 2012 Kindergarten Provision Census, which collected worker data from all service providers included in the WA August Student Census.

WA August Student Census

Institutional Environment
The organisation responsible for the School Census is WA Department of Education (DoE). Data for the WA August Student Census were collected under the School Education Act 1999 (WA).

Relevance
The data collected through the School Census by the WA DoE were used to meet national reporting requirements, including the National Schools Statistics Collection (NSSC) and the National ECEC Collection. The data were also used to determine allocation of schools operating grants and forward planning for the government and non-government school sectors, including Indigenous participation policy and ‘guaranteed local access to kindergarten’ policy.

Scope: The scope of the early childhood component of the WA August Student Census consisted of children aged 4 years on or before 30 June 2012, who were attending a preschool program within a school, including government and non-government schools, and community based preschools.

The scope for data provided on workers included all paid educators (teachers and education assistants) delivering preschool programs at all government, Catholic and independent schools, and community-based preschools.

Coverage: Full coverage was achieved for all registered and funded schools (both government and non-government) and community-based preschools.

As preschool programs provided in Long Day Care (LDC) settings are not funded or regulated by WA DoE, these service providers were not included in the WA August Student Census.

Data collection methods: The 2012 WA August Student Census collected data for Government and community-based preschools, and some non-government schools (using a URL data collection methodology) for the following statistical entities:

• Child: data at the child record level
  
• Service provider organisation: data at the service provider record level

• Worker: data at the worker record level.

Timeliness
The WA August Student Census is an annual collection conducted in August. In 2012 the collection was undertaken on 3 August with a reference period of 23 July to 3 August 2012. This reference fortnight occurred in the first two weeks of Semester 2, the first day of which was a pupil-free day in Western Australia. The 9-day fortnight for this particular census period was derived to a weekly equivalent to enable data comparison with other jurisdictions for the National ECEC Collection. Completion of the data processing and validation stage took approximately 2 to 3 months after the collection date, at which time data were delivered to the ABS.

Data sourced from the WA August Student Census for the purposes of the National ECEC Collection are published by the ABS in Preschool Education, Australia, 2012 (cat. no 4240.0).

Accuracy
Child level data: Government preschool data were obtained from government primary schools through an administrative system, the INTEGRIS Student Information Database (SID) and a daily data harvesting process. The SID is the primary source of URL data for public schools, which is loaded into the Student Census System data warehouse twice a year. Once collected, non-government school data are held in the Student Census System data warehouse. The census data is collected twice a year, in February and August. For non-government schools, child URL data is extracted from school enrolment records at the census date and made available to DoE. School principals are required to authorise census data submissions.

Worker level data: All educators (teachers and education assistants and Aboriginal and Islander Education Officers) who work with kindergarten children for more than one hour per week were required to provide details on their field and level of highest qualification relevant to ECEC. If an educator had no qualifications, they were required to indicate whether they were currently working towards an ECEC qualification and also the number of years of experience they had in teaching preschool (the first year of school, known as ‘kindergarten in WA, and offered part-time) or Pre-primary (the first year of full-time school in WA). To be registered as a teacher in WA, teachers are required to produce evidence that they are four year trained or equivalent. Equivalency is determined by WA College of Teaching (WACOT).

Government and non-government data collected through the Schools Census were then stored in the WA data repository, and extracted for the purposes of the National ECEC Collection.

Internal validation processes for the School Census occurred at three different points of the process. The Census module within SID had a series of in-built validation rules, some of which required the school to edit data before continuing, while other rules generated warnings alerting the user to anomalies. These warnings allowed the user to decide whether data needed amending before the census return was submitted. For government schools, a validation tool allowed administrators to bulk check the integrity of child date of birth data. This first level of validation only applied to WA government schools/preschools.

WA non-government schools used a comma-separated values file upload (URL data collection methodology) to the online Census System, which incorporated a number of internal consistency checks. A proportion of children in non-government managed preschools ages were imputed from aggregate data but supplied by WA DoE as pseudo URL child level data.
 
For non-government preschools, a validation checklist was utilised to ensure that:

• preschool children were counted by year of birth

• Aboriginal and non-Aboriginal data were correct

• totals were correct

A second set of validation processes were run by the WA August Student Census team, using both routine and ad-hoc validation methods to examine data while being assembled into working files. These processes were designed to enable effective investigation of new issues.

The third set of validation rules were built into the custom software which loaded school census data into the data warehouse, or into the staging phase.

Coherence
For the purposes of the National ECEC Collection, data collected from government school-based preschools using a URL data collection methodology were comparable with data collected in 2011.

Specific data elements collected using methods which did not align with the ECEC National Minimum Data Set (ECEC NMDS) specifications are outlined below and should be taken into account when analysing the data collected.

Attendance hours were constructed by applying average hours attended data. This data was available through student attendance data which became available in October. An average was calculated and applied for Indigenous and non-Indigenous data. Due to WA policy on enrolment and attendance there may be an under-count in children enrolled in preschool. This is because in WA a child attending for less than 90% of enrolled hours is not considered enrolled.

An analysis of the coherence of the WA collection with other jurisdictional collections should be undertaken on an individual basis, taking into account any data elements that are not consistent with the ECEC NMDS specifications. Deviations between the 2012 ECEC NMDS and data collected from the WA August Student Census, are outlined in table 6.5.

Data Element	Details of Deviation
Maximum preschool program hours available (Child)	This data element did not align with the ECEC NMDS as it was collected as minimum hours enrolled.
Preschool program hours attended (Child)	This data element was imputed from a separate WA attendance rates database.

Counts of Children:
In the collection reference period, it is possible for a child to be concurrently enrolled in and attending two or more preschool programs. In order to address collection requirements, it is important to be able to collate all matching child records.

Children in a Preschool Program in 2012 and in the Year Before Full-time Schooling:

For WA, counts of children enrolled in and attending preschool programs in 2012 were able to be derived (for most sectors, excluding non-government schools) as data were collected using a URL data collection methodology.

Interpretability
For government schools, the collection instrument itself contained significant guidance and computation rules to prevent inconsistent responses in relation to prior data entries. For non-government schools, a checklist was provided which included instructions and suggestions for validations. DoE staff who prepared the Census were trained to provide phone and email support through the DoE Customer Service Centre system.

Accessibility
Data sourced from the School Census for the 2012 National ECEC Collection are published at the state and territory level in Preschool Education, Australia, 2012 (cat. no 4240.0) which is available on the ABS website.

Information Source
The information contained within this report has been sourced from WA DoE in consultation with the ABS. Further information is available on the WA DoE website http://www.det.wa.edu.au/.

Previous Page

Next Page